python - 值错误: expected a dict or pandas.数据帧

Question

我正在尝试创建一个分类 vBar，它将显示为各种操作挑选的移民总数，但是，当我将“groupby”pandas 对象传递到列数据源时，我不断收到错误，并且我'我不太确定我做错了什么。

我在几个地方查找过类似的问题，但似乎找不到任何答案。

有人能指出我正确的方向吗？

#Imports
import pandas as pd
from bokeh.io import curdoc
from bokeh.plotting import figure
from bokeh.models import ColumnDataSource
from bokeh.models.tools import HoverTool
from bokeh.models import Button
from bokeh.layouts import row

#Global variables
viewFilter='Operation'

#Data
df = pd.read_csv('data.csv')
grouped = df.groupby(viewFilter)['Total Migrants']
source = ColumnDataSource(grouped)
#grouped = df.groupby(viewFilter)['Total Migrants'].sum()
#source = ColumnDataSource(pd.DataFrame(grouped))
operations = source.data[viewFilter].tolist()

# Fig Creation Function 
def create_figure():
    global viewFilter
    p=figure(x_range=operations)
    p.vbar(x=viewFilter, top='Total Migrants', 
           source=source, width=0.70)

    p.title.text='Demo Chart'
    p.xaxis.axis_label = viewFilter
    p.yaxis.axis_label = 'Total Migrants'

    #Hover took
    hover = HoverTool()
    hover.tooltips=[
        ("Total Migrants Rescued", "@{Total Migrants}")]
    hover.mode='vline'
    p.add_tools(hover)

    return p

#Update Data with Ship-level aggregation
def shipUpdate():
    print("Ship Button was Pushed")

#Widgets
shipButton = Button(label='Ship Level')
shipButton.on_click(shipUpdate)

#Implement Layout
layout = row(shipButton, create_figure())

#Add Layout to Document
curdoc().add_root(layout)

Answer 1

似乎如果我显式传入 pandas dataframe 对象，它就会解决此错误:

source = ColumnDataSource(pd.DataFrame(grouped))